arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2401.10754cs.LGcs.NI
收藏

面向流量分类的数据增强

Data Augmentation for Traffic Classification

Huawei Technologies SASU, France(华为技术法国有限公司);EURECOM, France(法国EURECOM)

Chao Wang, Alessandro Finamore, Pietro Michiardi, Massimo Gallo, Dario Rossi

AI总结:

本文针对流量分类中数据增强应用不足的问题,在3个数据集上以分组时间序列为输入基准测试18种增强函数,发现顺序与掩码增强效果更优,并可通过潜在空间分析解释其影响。

摘要:

数据增强(Data Augmentation,DA)——通过添加合成样本来丰富训练数据——是计算机视觉(CV)和自然语言处理(NLP)任务中广泛采用的一种技术,用于提升模型性能。然而,DA在网络场景中一直难以获得广泛应用,尤其是在流量分类(TC)任务中。在本研究中,我们填补了这一空白:以分组时间序列作为输入表示,并考虑多种训练条件,对应用于3个TC数据集的18种增强函数进行了基准测试。我们的结果表明:(i)DA能够获得此前未被探索的收益;(ii)作用于时间序列顺序和掩码的增强比幅度增强更适合TC;(iii)对基础模型的潜在空间进行分析,有助于理解增强对分类性能的正面/负面影响。

英文摘要:

Data Augmentation (DA) -- enriching training data by adding synthetic samples -- is a technique widely adopted in Computer Vision (CV) and Natural Language Processing (NLP) tasks to improve models performance. Yet, DA has struggled to gain traction in networking contexts, particularly in Traffic Classification (TC) tasks. In this work, we fulfill this gap by benchmarking 18 augmentation functions applied to 3 TC datasets using packet time series as input representation and considering a variety of training conditions. Our results show that (i) DA can reap benefits previously unexplored, (ii) augmentations acting on time series sequence order and masking are better suited for TC than amplitude augmentations and (iii) basic models latent space analysis can help understanding the positive/negative effects of augmentations on classification performance.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑